MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202631085758 A) filed by Narula Institute Of Technology on July 13, 2026, for “adaptive Hybrid- Scvmnn Handwritten Recognition Framework With Dynamic Preprocessing And Edge Deployment”.
Inventors include Mr. Abhishek Banerjee; and Ms. Debrupa Pal.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: “Adaptive Hybrid CNN-SVM Handwritten Recognition Framework with Dynamic Preprocessing and Edge Deployment ” ABSTRACT The present invention relates to an Adaptive Hybrid CNN-SVM Handwritten Recognition Framework with Dynamic Preprocessing and Edge Deployment designed to provide accurate, scalable, and computationally efficient handwritten digit and character recognition for real-time applications. The proposed framework integrates the automatic hierarchical feature extraction capability of Convolutional Neural Networks (CNNs) with the optimized classification efficiency of Support Vector Machines (SVMs) to improve recognition accuracy while reducing computational overhead and inference latency. The invention introduces a dynamic preprocessing engine capable of automatically adapting to varying handwriting styles, image distortions, stroke thickness variations, illumination conditions, incomplete character structures, and noisy environments. (Figure 2)
Disclaimer: Curated by HT Syndication.
